Every hadith has two parts. The isnad is the chain of transmission, a list of names running back from the compiler to a Companion who heard or saw the thing reported. The matn is the text itself, usually a few lines. The classical science concentrates overwhelmingly on the chain, on the reasoning that a report can only be as good as the people who carried it.
That produced rijal, the biographical evaluation of narrators: vast dictionaries recording when each transmitter lived, whom he studied under, whom he taught, whether he was truthful, whether his memory decayed in old age, whether he was known to have met the person he claims to quote. From this comes the grading. A sahih report has a continuous chain of upright and precise narrators with no hidden defect; hasan is sound but with a weaker link; daif is weak, through a break in the chain or a discredited narrator; mawdu is judged fabricated outright. A separate axis counts transmission lines: mutawatir reports come through so many independent chains that collusion is held impossible, while the great majority are ahad, transmitted singly, and jurists disagree about how much doctrinal weight those can bear.
The grading is a working tool, not a historical verdict everyone shares, and it is applied with judgement. Weak hadith are not simply discarded, many scholars permit their use in exhortation while refusing them in law. Different scholars have graded the same report differently, and grading disputes are a live part of Islamic scholarship rather than a settled inheritance.